© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 48539
    Prev
    Next
    INDEX
    Explanations

    legal citations and numbering

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects the structured docket‐and‐citation header elements typical of legal opinions (case names, court names, docket numbers, and dates).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Utah
    -1.64
     Colorado
    -1.62
    Utah
    -1.47
    Idaho
    -1.45
     Idaho
    -1.45
    Colorado
    -1.42
     Denver
    -1.21
     Montana
    -1.17
     Mesa
    -1.14
     Wyoming
    -1.13
    POSITIVE LOGITS
     Clim
    0.97
     Wabash
    0.94
     mangrove
    0.93
     impati
    0.93
     Champaign
    0.91
     beagle
    0.90
     Tamil
    0.85
    Tamil
    0.84
     Gloucester
    0.84
     Ithaca
    0.84
    Activations Density 0.032%

    No Known Activations